History and Role of Policing

Before there was a police force

English policing grew out of shared obligation rather than paid employment. Households were grouped into tithings of ten families, tithings into hundreds, and hundreds into shires supervised by a shire-reeve, ancestor of the American sheriff. The constable, appointed locally and responsible for the hundred's weapons, is the earliest figure the 1967 President's Commission on Law Enforcement identifies as a police officer. Under Edward I (1272-1307) larger towns acquired a 'watch and ward' that guarded gates and arrested night-time offenders, and in 1326 Edward II created the office of justice of the peace; the constable became the justice's subordinate, a hierarchy that held for roughly five centuries.

By the eighteenth century the arrangement had decayed. Service everyone owed became service nobody performed: constables paid substitutes to take their turns, and the Commission describes those substitutes as badly paid and often too old to be effective. Night watchmen nominally guarded cities against thieves but reportedly did little beyond calling out the hour. Industrializing London improvised: government created public offices, later called police offices, each with three paid justices authorized to employ six paid constables, and civic bodies formed patrols such as the Bow Street Horse and Foot Patrol. By 1800 nine such offices existed in metropolitan London, and the Commission records that they largely refused to communicate with one another for fear of losing credit for an arrest. That is what 1829 was meant to solve: not the absence of law enforcement, but the absence of a coordinated, full-time, publicly accountable one.

1829, and the principles Peel probably did not write

Robert Peel became Home Secretary in 1822 arguing that better policing could not eliminate crime but that bad policing was feeding disorder. Seven years later he steered through Parliament an Act for improving the police in and near the metropolis, recorded by legislation.gov.uk as the Metropolitan Police Act 1829, 1829 c. 44 in the tenth year of George IV. The force numbered about a thousand uniformed men under two magistrates, later titled commissioners, who held administrative but deliberately not judicial power, answering to the Home Secretary and through him to Parliament. An 1839 statute let county justices establish forces; an 1856 statute required every borough and county to have one.

The famous nine 'Peelian principles' are a separate matter, and their provenance is genuinely disputed. The Home Office's own published note on policing by consent states that there is no evidence of any link to Robert Peel, that the principles were likely devised by the first Commissioners of Police of the Metropolis, Charles Rowan and Richard Mayne, and that they were set out in the General Instructions issued to every new officer from 1829. Keep the distinction clean: the ideas the list expresses - that the police are the public, that cooperation falls as force is used, that the test of policing is the absence of crime rather than visible action - are real and old. The claim that Peel wrote them is the part that does not survive checking.

Two American lineages, and the argument between them

Colonists imported constables and sheriffs largely unchanged. Boston had night watchmen by 1636; New York's carried rattles and were known as the Rattlewatch. As cities grew, the watch failed the way the English one had. Boston created a day force in 1838 alongside its watch. The New York legislature passed a law in 1844 authorizing the first unified day-and-night police and abolishing the watch; the standard list dates the resulting department to 1845. Boston consolidated in 1854. Hubert Williams and Patrick Murphy, writing in the same NIJ series, reproduce the conventional sequence of 'new police' dates, which begins Boston 1838, New York City 1845 and Chicago 1851 and runs on to Detroit in 1865.

Their central point is that this list is not the whole country. Southern colonial legislatures had enacted slave patrol statutes decades earlier - South Carolina in the 1740s, Georgia in 1757 - creating compulsory patrols empowered to search the dwellings of enslaved people for weapons, punish anyone found off a plantation without a pass, and enter houses suspected of harboring or trading with them. Charleston, Savannah and Richmond ran combined foot and mounted patrols, and in Charleston policing was the largest single item in the municipal budget. Williams and Murphy report, attributing the argument to the historians Samuel Walker and Philip Reichel, that a strong case exists that the first American modern-style policing occurred in the slave patrols.

Hold that claim's shape precisely, because it is easy to overstate in either direction. It is documented that specific southern jurisdictions built permanent, publicly authorized, continuously patrolling forces for racial control before northern cities built municipal departments, and that there the patrol is a real institutional ancestor. It is an interpretive argument, not a settled finding, that this makes slave patrols the origin of American policing as a whole; Boston's and New York's departments have their own traceable descent from the watch. Williams and Murphy also note that northern departments were created amid anti-abolitionist and anti-Catholic rioting in the 1830s and 1840s, so even the northern story is more about controlling disorder than detecting crime. Historians differ over how much weight each strand carries, and a lesson that resolves that disagreement for you is doing you a disservice.

Three eras, and the criticism of the periodization

The standard map of American police history is the one George Kelling and Mark Moore published for NIJ in November 1988. Their political era, from the introduction of police into municipalities in the 1840s to the early 1900s, drew authority from ward politicians, defined the function broadly enough to include soup lines and lodging for newly arrived immigrants, ran through semi-autonomous precincts on foot patrol, and measured success as political and citizen satisfaction. Its weaknesses were corruption, curbstone justice against outsiders, and almost no supervisory control.

The reform era took hold in the 1930s, peaked in the 1950s and 1960s, and eroded in the late 1970s. August Vollmer, chief in Berkeley, rallied executives around reform in the 1920s and 1930s, and his protege O.W. Wilson became its principal administrative architect. Reformers replaced political authorization with law and professionalism, used civil service to break patronage hiring, narrowed the function to crime control, and organized departments on Taylorist lines. Wilson supplied the theory that marked cars patrolling randomly would create a sense of police omnipresence, and departments pushed centralized dispatch and 911, marketing rapid response as effective crime control. The 1967 President's Commission ratified the shift by recasting police as part of a criminal justice system rather than an agency of city government. The third era, community problem solving, widens the function again toward order maintenance and problem solving.

Kelling and Moore themselves warn that the boundaries are not clean. The sharper criticism came from Williams and Murphy in 1990, in the same series, who called the analysis 'disturbingly incomplete': it takes little account of slavery, segregation and discrimination, treats departments as autonomous strategists rather than as barometers of the society around them, and generalizes from the urban Northeast. Use the three eras as a filing system for changes in authorization, function, organization and tactics, not as a chronology every department passed through on schedule.

The studies that dismantled the reform-era assumptions

Three studies did more than any argument to unsettle the professional model, and each is routinely misquoted, so specifics matter. The Kansas City Preventive Patrol Experiment ran from 1 October 1972 to 30 September 1973 in fifteen South Patrol Division beats, matched on crime, calls for service, ethnic composition, median income and transiency, then split into five reactive beats with no preventive patrol, five control beats at the usual one car per beat, and five proactive beats at two to three times that level. Across victimization surveys and reported crime the three conditions showed no significant differences in crime, citizen attitudes, fear, response time, or satisfaction with response time. The report's cautions matter as much: it did not show visible presence can never matter in selected circumstances, it warned against reading it as a case for fewer officers, and it noted officers' noncommitted time ran to sixty percent.

Peter Greenwood and Joan Petersilia's RAND study, report R-1776-DOJ (1975), concluded that traditional investigative approaches do not significantly affect how often cases are solved, that most are solved by routine administrative procedures, that roughly half of investigative effort could be eliminated or redirected without unduly reducing effectiveness, and that apprehension rates are likelier to rise from alert patrol and citizen cooperation than from better detective work. RAND proposed restructuring investigation, including a major offenders unit, not abolishing it.

Rapid response was the third assumption to go. William Spelman and Dale Brown's 1981 study for the Police Executive Research Forum examined more than four thousand victims and witnesses in four cities and found that citizen reporting time, not police response time, most affects the chance of an on-scene arrest: shaving minutes off dispatch pays when a crime is reported as it happens and little otherwise, since most calls arrive after the offender has gone. None of these studies says police do not matter. Each says one tactic delivered less than its reputation promised.

What the job is actually made of

The most careful early measurement is Eric Scott's 1981 NIJ study, in which observers coded 26,418 citizen calls in departments across Rochester, St. Louis and Tampa-St. Petersburg into twelve categories, recording each as the caller described it. Requests for information were the largest category at twenty-one percent, nonviolent crime second at seventeen, then assistance at twelve, public nuisances at eleven, traffic at nine, interpersonal conflict at seven, and violent crime at two. Fewer than a fifth of calls concerned criminal incidents at all. Two cautions travel with those numbers: departments classify the same event differently, so category boundaries move the percentages, and the familiar claim that about eighty percent of patrol workload is noncriminal inherits that ambiguity even though studies converge on it.

Contemporary data show the same shape without settling on one number. Cynthia Lum, Christopher Koper and Xiaoyun Wu analyzed millions of 911 calls across nine U.S. agencies in a 2022 Police Quarterly article and reported that the range and volume of what people call police about is very large and mostly not obviously transferable to other organizations without substantial added resources. The Bureau of Justice Statistics found from its 2022 Police-Public Contact Survey, published in October 2024, that nineteen percent of U.S. residents aged sixteen or older, about 49.2 million people, had police contact in the previous twelve months: 7.6 percent police-initiated, 11.2 percent resident-initiated, 2.5 percent from a traffic accident. Fifty-eight percent of those whose most recent contact was police-initiated were drivers in traffic stops, and forty-six percent of resident-initiated contacts were to report a possible crime. The modal encounter between an American and an officer is a traffic stop; the modal reason an American contacts police is to report something. That is a regulatory and service relationship far more often than an adversarial one - the conclusion the 1967 Commission had already reached when it wrote that officers perform a wide range of complex functions beyond serious crime.

Legitimacy, procedural justice, and stating the evidence honestly

The dominant contemporary framework for the police role is procedural justice, developed principally by the social psychologist Tom Tyler. Its claim is that people cooperate with legal authority mainly because they regard it as legitimate, and judge legitimacy from how they are treated rather than from whether they liked the outcome. The National Academies committee that reviewed proactive policing in 2018 sets out four elements: voice, the chance to state your side before a decision; neutrality, consistent and explained rule-based decisions; respect, dignified treatment; and trustworthiness, a belief that the officer's motives are benevolent.

Report that evidence at the strength it has. The committee found perceptions of procedurally just treatment strongly associated with judgments of legitimacy and willingness to cooperate, but concluded the research base was insufficient to establish that procedurally just policing causally produces either (Conclusion 6-4). On crime it found the evidence base very small: too small to conclude that procedural justice policing affects crime or disorder, and too small to conclude it does not (Conclusion 4-11). Insufficient evidence is not evidence of absence, the committee stressed, and procedural justice has stronger causal support in courts and workplaces. So: a well-supported association, a promising theory, an open causal question in policing.

Two habits carry this topic: separate event, evidence and interpretation - 1829 is an event, the Home Office note is evidence about authorship, 'policing by consent' is an interpretation - and date every finding while naming its jurisdiction. This lesson is educational material on institutional history, not legal advice, and takes no position on current policy. Police discretion, use of force, community policing and search and seizure law have their own lessons in this unit.